Paying a price: Culture, trust, and negotiation ... - Brian Gunia

Download PDF
4 downloads 474 Views 376KB Size Report
Comment
deep-seated cultural beliefs about trust prepare Indian managers to negotiate successfully in the global economy, where complex problems often get resolved  ...
1 Running Head: Culture, Trust, and Negotiation

Paying a price: Culture, trust, and negotiation consequences

Brian C. Gunia Jeanne M. Brett* Northwestern University Kellogg School of Management

Dishan Kamdar Amit Nandkeolyar Indian School of Business

Journal of Applied Psychology, 2011

© 2010 American Psychological Association. http://www.apa.org/pubs/journals/apl/index.aspx. This article may not exactly replicate the final version published in the APA journal. It is not the copy of record.

Culture, Trust, and Negotiation 2

Paying a price: Culture, trust, and negotiation consequences

Abstract Three studies tested hypotheses predicting that Indian and American negotiators‘ propensity to trust accounts for their strategy, insight, and joint gains in negotiation. Study 1, a survey, documented that Indian negotiators trust their counterparts less than American negotiators. Study 2, a negotiation simulation, linked American and Indian negotiators‘ self-reported trust and strategy to their insight and joint gains. Study 3 replicated and extended Study 2 using independently-coded negotiation strategy data, allowing for stronger causal inferences. Overall, the strategy necessitated by Indian negotiators‘ low propensity to trust produced relatively poor outcomes. Our data support an expanded theoretical model of negotiation, linking culture and trust to strategies and outcomes. Our discussion suggests an approach that may help low-trust negotiators in general perform better.

Key words: negotiation, trust, culture, joint gains, India

Culture, Trust, and Negotiation 3 Introduction After a decade of explosive growth, India became the world‘s third-largest economy in 2006 (OECD, 2007). With outsourcing revenue approaching $50 billion (Reuters, 1/29/08) and a $44 billion merchandise trade with the U.S. alone (U.S. Census Bureau, 2009), India is already playing a major role on the world‘s economic stage. Still, 80% of India‘s population lives on less than $2 a day (UNDP, 2008), 50% has no access to electricity (The Economist, 12/11/08), and the country is 84th on the 2009 corruption perceptions index (Transparency International). Thus, India has been and continues to be a culture in the throes of economic development. As India‘s development continues, business negotiations within India and across cultures will increasingly shape whether the country stagnates or integrates into the world economy (Kumar & Worm, 2004). If Indian managers can use negotiations to create joint gains (value that benefits both themselves and their counterparts), deals should close and businesses prosper – improving Indian prospects and propelling the world‘s growth. If Indian managers negotiate poorly – reaching deals that leave potential gains on the table – India‘s businesses will suffer, with consequences for its billion-plus citizens and the world economy as a whole. This paper addresses an applied question with theoretical importance: whether India‘s deep-seated cultural beliefs about trust prepare Indian managers to negotiate successfully in the global economy, where complex problems often get resolved through integrative negotiations – negotiations involving the creation of joint gains (Adair, Weingart, & Brett, 2007; Brett, 2007; Requejo & Graham, 2008). Prior research on integrative negotiations has established a theoretical model linking negotiation strategy to insight, and insight to joint gains (e.g., Pruitt & Lewis, 1975; Weingart, Thompson, Bazerman, & Carroll, 1990). Our theorizing adds an exogenous variable (culture) and an endogenous variable (trust) to the front of this model (see

Culture, Trust, and Negotiation 4 Figure 1). We test our model with three studies. Study 1 uses survey methods to document Indian and American MBA students‘ cultural differences in trust propensity during negotiation. Study 2 engages Indian and American managers in a negotiation simulation and uses their negotiated outcomes and post-negotiation questionnaire data to demonstrate relationships among the variables in Figure 1. Because Study 2‘s survey data preclude causal inference, Study 3 uses the same simulation among different samples of Indian and American managers to test the causality of the model‘s relationships, with data from coded negotiation transcripts. India, a culture with Eastern and Western influences, in the throes of economic transition, provides a particularly clear look at the relationships in our model. Across these three studies, we document that Indian managers‘ trust propensity in negotiations may ill-equip them for achieving joint gains. Compared to American managers, we find that: 1) Indian managers bring a lower propensity to trust into (and out of) negotiations; 2) this lower trust propensity leads Indian managers to rely on a negotiation strategy of substantiation and offers, rather than questions and answers; 3) the use of substantiation and offers undercuts Indian managers‘ achievement of the relevant tradeoffs; and 4) Indian managers realize substantially lower joint gains than American managers – leaving an average of $1.7 million, versus $0.98 million, on the table. From Negotiation Strategy to Joint Gains Most negotiations that managers face are not zero-sum (Bazerman & Neale, 1992; Brett, 2007). Rather, they present opportunities to create joint gains. Creating joint gains serves both parties‘ self-interest: It increases the resources each can claim; facilitates agreement by expanding, or even creating, a zone of possible agreement; and promotes stable agreements and long-term relationships (Brett, 2007; Kimmel, Pruitt, Magenau, Konargoldband, & Carnevale, 1980; Pruitt, 1981; Pruitt & Lewis, 1975; Raiffa, 1982; Walton & McKersie, 1965; Weingart et

Culture, Trust, and Negotiation 5 al., 1990). Creating joint gains is thus one of the most important objectives in negotiation (Brett, 2007; Kimmel et al., 1980; Pruitt & Lewis, 1975; Raiffa, 1982; Requejo & Graham, 2008). Opportunities for joint gains arise when negotiators attach different priorities to the issues. Creating joint gains involves reaching ―insights,‖ i.e., discovering the tradeoffs that give negotiators favorable terms on their highest-priority issue(s) (Kimmel et al., 1980), and incorporating those insights into agreements (Pruitt, 1981; Raiffa, 1982). Negotiators achieve or fail to achieve insights and joint gains by using strategies: sets of actively- or passively-chosen, goal-directed behaviors (Weingart et al., 1990). The most consistently-documented strategy for realizing insight and joint gains is to ask questions and provide answers early in the negotiation, and later integrate the revealed information into offers (Brett, 2007; Kimmel et al., 1980; Olekalns & Smith, 2003; Pruitt & Lewis, 1975; Weingart et al., 1990). For our purpose, ―questions‖ are interrogative statements made to elicit informationsharing, and ―answers‖ connote any information-sharing, even when no explicit question is asked (e.g., Kimmel et al., 1980). Since the norm of reciprocity (Gouldner, 1960) eventually requires those who ask questions to answer them, questions and answers constitute a single strategy: ―Q&A (Adair & Brett, 2005). Because it involves a logical and sequential exchange of information, Q&A seems to work especially well for negotiators who use low context communication (Adair & Brett, 2005) and its associated linear versus holistic thought patterns, rational versus spiraling arguments, and direct versus indirect meanings (Glenn & Glenn, 1981; Hall, 1976; Nisbett, Peng, Choi, & Norenzayan, 2001). Ultimately, low context negotiators incorporate information gleaned from Q&A into offers (Adair & Brett, 2005). Yet, not all low context negotiators rely on Q&A before making offers. Some use offers, and the associated substantiation (persuasion) attempts, from the

Culture, Trust, and Negotiation 6 negotiation‘s outset (Kimmel et al., 1980). However, these negotiators tend to miss the relevant tradeoffs and realize poor joint gains (Adair et al., 2007; Kimmel et al., 1980; Pruitt, 1981; Pruitt & Lewis, 1975; Weingart et al., 1990). American negotiators who exchange substantiation and offers early in negotiations tend to get locked in positional, issue-by-issue haggling and miss insights into mutually-beneficial tradeoffs (Adair, et al., 2007). Thus, Q&A seems well-suited to generating joint gains, while substantiation and offers (S&O) tends to undermine joint gains. From Trust to Negotiation Strategy Many scholars have commented on the link between negotiation strategy and trust (Butler, 1995, 1999; Deutsch, 1973; Kimmel et al., 1980; Walton & McKersie, 1965; Zand, 1972). In general, trust is ―a psychological state comprising the intention to accept vulnerability based on positive expectations of the intentions or behavior of another‖ (Rousseau, Sitkin, Burt, & Camerer, 1998: 395). In negotiation, ―another‖ is a counterpart, and vulnerability stems from the counterpart‘s ability to exploit information that a negotiator shares – i.e., to take advantage (Butler, 1999). High-trust negotiators believe that counterparts will use shared information in good faith; low-trust negotiators are not sure what they will do. The Q&A strategy requires trust because both questions and answers give the counterpart an opportunity to take advantage (Butler, 1999). Questions invite vulnerability by revealing gaps in a negotiator‘s knowledge. They also set the norm of reciprocity (Gouldner, 1960) in motion, increasing the likelihood that askers will eventually be asked themselves. Answers create vulnerability because they often reveal sensitive information about a negotiator‘s private preferences (Kimmel et al., 1980; Pruitt & Lewis, 1975). Vulnerabilities notwithstanding, negotiators need to understand each other‘s preferences to appreciate tradeoffs and achieve joint gains. High trust in negotiations – an intention to accept vulnerability based on the belief that

Culture, Trust, and Negotiation 7 counterparts will use shared information to identify mutually-beneficial, rather than selfinterested opportunities (Kimmel et al., 1980) – enables them to surface preferences via Q&A. Low trust casts Q&A as unwise at-best, and an invitation to take advantage at-worst. Seen as the reluctance to accept vulnerability based on dubious beliefs about the counterpart, low trust changes the calculus of Q&A. If asking questions reveals incomplete knowledge, and the counterpart may answer deceitfully, why ask at all? If answering reveals private preferences, and the counterpart may exploit them, why answer at all? Thus, low-trust negotiators are likely to fall back on behaviors that reduce their vulnerability: S&O (Kimmel et al., 1980). Neither substantiation nor offers requires trust, because neither reveals much about a negotiator (Adair et al., 2007; Kimmel et al., 1980; Pruitt & Lewis, 1975). Rather than windows into a negotiator‘s private knowledge or preferences, S&O are readily interpreted as the fulfillment of fixed-pie expectations (Bazerman & Neale, 1992; Fisher & Ury, 1981). Thus, Q&A strategy should stem from high trust and S&O from low trust. From Culture to Trust People from different national cultures vary in their willingness to trust (Ferrin & Gillespie, 2010; Inglehart, Basáñez, & Menéndez Moreno, 1998; Johnson & Cullen, 2002). Across many interpersonal interactions, Westerners (i.e., North Americans, Western Europeans) tend to make the ―swift trust‖ assumption: others deserve to be trusted until they prove otherwise (Dirks, Lewicki, & Zaheer, 2009; Meyerson, Weick, & Kramer, 1996; Weber, Malhotra, & Murnighan, 2005). Easterners (i.e., East and South Asians) generally trust less than Westerners (Delhey & Newton, 2005; Yamagishi & Yamagishi, 1994), but also condition their trust on the situation (Branzei, Vertinsky, & Camp, 2007).

Culture, Trust, and Negotiation 8 Socio-political data imply that the trust propensity of American and Indian managers should reflect these broad, regional trends: The U.S. ranks 65 places higher than India (19 versus 84) on Transparency International‘s 2009 corruption perceptions index (CPI), reflecting lower domestic and international trust in Indian than American institutions. Additionally, Indians reported a lower generalized trust than Americans on many questions of the World Value Survey (WVS). For example, Indians agreed that ―Most people would try to take advantage of you,‖ while Americans disagreed (Inglehart et al., 1998). Additionally, Indians registered as less trusting than Eastern neighbors like China and Japan on both the CPI and WVS. By comparison, cross-cultural research (e.g., Buchan, Croson, & Dawes, 2002; Huff & Kelley, 2003) identifies the U.S. as one of the most trusting countries. Although trust varies by culture in general, it is not clear whether trust varies by culture in particular situations, like negotiations. Indeed, Easterners‘ sensitivity to situational context (Branzei et al., 2007) suggests that general patterns may not hold across situations. Our goal is to examine differences between American and Indian managers‘ propensity to trust in negotiation situations, and the consequences of trust for negotiation strategy and outcomes. Since negotiation is a situation stereotypically associated with self-interest (Bazerman & Neale, 1992; Branzei et al., 2007; Fisher & Ury, 1981), especially in Eastern cultures (Druckman, Benton, Ali, & Bagur, 1976; Fang, 1999; March, 1988), we propose that negotiation may exacerbate Indians‘ cultural propensity for low trust. In contrast, Americans are not particularly sensitive to situational context (Branzei et al., 2007), so it seems likely that they will import swift trust (Meyerson et al., 1996) into negotiations. If our reasoning is correct, trust in negotiation should follow discernable patterns: Americans, applying swift trust, should trust counterparts until they prove untrustworthy.

Culture, Trust, and Negotiation 9 Indians, characterizing the negotiation as a self-interested context, should not trust their counterparts until they prove trustworthy. Thus, in negotiations, culture and context should act like individual trust propensity (―a stable individual difference that affects the likelihood that a person will trust‖) (Colquitt, Scott, & LePine, 2007: 910): Hypothesis 1: Indian negotiators will exhibit a lower trust propensity than American negotiators. Although we expect Indian and American negotiators to differ in trust propensity, we do not expect them to define the underlying concept of trust differently. If they did, it would imply a difference in conceptualization rather than substance. Thus, our first study also investigates how Indian and American negotiators define trust in negotiations. Given the ―no difference‖ nature of our prediction, we do not formulate it as a formal hypothesis. Study 1: Culture and Trust Propensity in Negotiations Study 1 was a Web survey, administered to matched samples of American and Indian MBA students. It was designed to test Hypothesis 1, corresponding to the culture-trust relationship in Figure 1, and to investigate Indian and American negotiators‘ definitions of trust. The two institutions from which our samples were drawn (here and in future studies) are similar in many respects: both are top-tier, globally-ranked business schools located outside of cities with populations near 4 million. Both have MBA and executive-level programs conducted exclusively in English, and both attract their own country‘s top managerial prospects, as well as many international students. In all studies, Indian and American participants were exposed to the same course material (if any) before data collection. Methods

Culture, Trust, and Negotiation 10 Procedures. Study 1 used a closed-ended Web survey of Indian and American MBA students enrolled in a negotiation strategies course. Approximately one week before the start of class, students received an email from their professor requesting participation. The email emphasized that participation was optional, promised personalized feedback via email, and provided a link to the survey. Email addresses and feedback were not available to the professor. Participants. 143 MBA students at a U.S. and 135 students at an Indian business school completed the survey, yielding response rates of 87.73% and 76.70%, respectively. From this pool, we retained respondents who reported that their nationality was American (Indian), and that their dominant culture was the same. The Indian sample was larger, younger, and more male, so we created matched samples by dropping the youngest 49 Indian, male respondents. Our final sample included 78 American and 78 Indian respondents, statistically equivalent in age and gender. All results were identical for the full sample of Indian respondents. The demographic characteristics of the final samples were: American age 28.72 (SD = 2.20), Indian age 28.38 (SD = 1.98), t(147) = 0.98, p = 0.33 (seven Indian participants declined to report); American gender 48.72% male, Indian gender 62.82% male, χ21 = 3.14, p = 0.08. Data and Analysis. We measured the independent variable, national culture, by the school where the data were collected and the self-reported nationality and culture of participants. We asked five, closed-ended questions using 7-point, Likert-type scales, 1=strongly disagree to 7=strongly agree to measure trust propensity in negotiations. Because our focus was trust propensity, we selected two questions tapping into that concept from a validated scale on trust in negotiation (Lewicki, Stevenson, & Bunker, 1997; Olekalns, Lau, & Smith, 2007): ―The other party will try to be someone who keeps promises and commitments‖ and ―The other party will do what they say they will do.‖ For reliability, we then wrote three more questions about

Culture, Trust, and Negotiation 11 propensity to trust. They were: ―In negotiations, most other parties are basically honest‖; ―In negotiations, there is no point in trusting the other party until the two of you have had repeated interactions‖; and ―In negotiations, you should not trust the other party, even if you know them well in other contexts.‖ Responses to the five questions were correlated, so we recoded them as appropriate and collapsed them into a trust propensity index (α = 0.71). To determine whether Indian and American negotiators defined the concept of trust similarly, we asked three questions (using 7-point scales, 1=not at all to 7=very much so) about whether trust in negotiation means ability, benevolence, and integrity (Mayer, Davis, & Schoorman, 1995) (see Appendix A). We expected no cultural differences. Results and Discussion Indian and American negotiators defined trust similarly, but, as predicted by Hypothesis 1, reported different propensities to trust. Respondents from both cultures agreed that trust means that the other party has ability (American M = 5.22, SD = 1.03; Indian M = 5.09, SD = 1.19), t(153) = 0.71, p = 0.48; benevolence (American M = 3.83, SD = 1.39; Indian M = 4.11, SD = 1.32), t(151) = 2.13, p = 0.21; and integrity (American M = 5.33, SD = 1.10, Indian M = 5.33, SD = 1.14), t(153) = 0.00, p = 1.00. These results suggest that Indian and American negotiators had the same construct in mind when thinking about trust. However, as predicted by Hypothesis 1, Indians‘ propensity to trust (M = 4.14, SD = 0.65) was significantly lower than Americans‘ (M = 4.50, SD = 0.60), t(154) = 3.64, p < 0.001. Gender and age had no effects. Study 1 suggested that Indian and American negotiators attach the same meaning to trust, but that the Indians bring a lower trust propensity into negotiations. This study lent initial support to our theorizing about cultural differences in the propensity to trust in negotiations. Study 1

Culture, Trust, and Negotiation 12 addressed negotiations in general, not a particular negotiation, and measured negotiators‘ beliefs, not their behaviors. Study 2 focused on behavior in a particular negotiation. Study 2: Culture, Trust, Reported Strategy, and Outcomes Study 2 proposes and tests hypotheses implied by each of the links in Figure 1. The first link is culture to trust. Study 2 elaborates on Hypothesis 1 by investigating trust in negotiation prior to a simulated negotiation. The second link connects trust and negotiation strategy. Study 2 proposes and tests hypotheses about the use of Q&A and S&O. The third link (connecting strategy and insight), and the fourth (connecting insight and joint gains) are also tested. Finally, Study 2 proposes and tests a feedback loop (illustrated in Figure 1) suggesting that experiences in one negotiation will influence trust in subsequent negotiations. We expected Study 2 data to confirm Hypothesis 1 and provide evidence of the strategic consequences of cultural differences in trust propensity. In the introduction to this paper, we suggested that a high trust propensity should facilitate Q&A strategy, since high-trust negotiators expect their counterparts to use information in good faith. We also suggested that a low trust propensity should facilitate S&O strategy, since low-trust negotiators hesitate to share or believe information. Trust propensity in negotiations, then, should predict negotiation strategy, as reflected in post-negotiation perceptions of own and counterpart behavior: Hypothesis 2a: American negotiators will report using and witnessing more behaviors indicative of Q&A strategy than Indian negotiators. Hypothesis 2b: Indian negotiators will report using and witnessing more behaviors indicative of S&O strategy than American negotiators.

Culture, Trust, and Negotiation 13 We also suggested in the introduction that Q&A would facilitate insight and joint gains, while S&O would undermine both1. If Indian negotiators resort to S&O more often than American negotiators, they should reach fewer insights and generate lower joint gains: Hypothesis 3: Indian negotiators will identify the relevant tradeoff opportunities less accurately than American negotiators. Hypothesis 4: Indian dyads will generate lower joint gains than American dyads. We state Hypothesis 4 at the dyadic level to emphasize that joint gains—unlike culture, trust, strategies, or insight—are only defined at the dyadic level. Finally, we expected that the foregoing strategies and outcomes would influence the trust that actually developed during the negotiation. This prediction rests upon prior research suggesting that initial trust is reaffirmed through behavioral confirmation – both outside (e.g., Colquitt et al., 2007) and inside (Olekalns & Smith, 2005) of negotiations. Thus, we propose that S&O, minimal insight, and low joint gains will reinforce cultural differences in trust propensity, resulting in a sustained cultural difference even after the negotiation. Thus:

1

We fully recognize that several studies have found that Japanese and other negotiators from high context

communication cultures can use S&O to reach insight and joint gains comparable to those of Western negotiators using Q&A (Adair et al., 2003; Adair & Brett, 2005, Adair, Weingart, & Brett, 2007). For Indian negotiators to use S&O in this fashion, they would have to be high context and holistic thinking, like the Japanese. The evidence is that Indians naturally use a low context, linear style of problem solving more akin to Westerners than Easterners (Nisbett, Peng, Choi, & Norenzayan, 2001), perhaps because of their geographic isolation from the East and substantial contact with the West (Kumar & Worm, 2004). Several empirical studies have documented that Indians communicate (Kapoor, Hughes, Baldwin, & Blue, 2003) and think (Allinson & Hayes, 1988; Hayes & Allinson, 2000; Parikh, Neubauer, & Lank, 1994) in ways at least as linear as Americans. Thus, we adopted the working premise that the consequences of Indians‘ and Americans‘ use of S&O would be more alike than different.

Culture, Trust, and Negotiation 14 Hypothesis 5: Controlling for trust propensity, Indian negotiators will report lower trust than American negotiators at the end of the negotiation. Methods Participants. Study 2 used executive samples, lending generalizability to our overall research program. Indian managers (N=56) were participants in one or the other of two executive programs at the same Indian business school from which Study 1‘s MBA sample was drawn. Across both data collections, the average age was 41.98, SD = 7.96 and the sample was 98.20% percent male. American managers (N=78) were participants in one of four Executive MBA classes at the same U.S. business school from which Study 1‘s MBA sample was drawn. The average age of the American sample was 37.94, SD = 5.79 and the sample was 77.60% percent male. All participants reported their gender, but 21 (spread across the two cultures) declined to provide their age. As in Study 1, managers had to indicate that both their nationality and their dominant culture was American (Indian) to qualify. All data were collected in the same year. There were more female negotiators in the American sample than the Indian sample, χ21 = 11.60, p = 0.01; however, no dyad in either sample consisted of two female negotiators. The Americans were also significantly younger (M = 37.94, SD = 5.79) than the Indians (M = 41.98, SD = 7.96), t(109) = 3.08, p = 0.01. Whereas Study 1 used matched samples, we controlled for demographic effects in this study, primarily to see if demographics had an effect. Ultimately, gender and age were not significant in any analyses. We randomly assigned participants to roles and dyads to minimize the chance that they knew one another. In one of the Indian samples participants did, and in the other sample they did not know each other before data collection. This difference within the Indian dataset provided a natural experiment to determine whether familiarity might generate trust and joint gains among

Culture, Trust, and Negotiation 15 Indian negotiators. It did not: there were no differences or trends in the dependent variables between the two groups of Indian managers. American dyads were constructed such that negotiators came from different classes and did not know one another. Simulation. All managers negotiated the Cartoon simulation (Dispute Resolution Research Center, 2008) either representing a buyer (a television station), or a seller (a film company). They negotiated over the sale of rerun rights for a cartoon series. They had to resolve the price of the cartoon (a distributive issue), and two tradeoff issues: the number of runs (how many times each of the 100 episodes could be shown during the fixed, five-year contract), and financing (how soon the money would be paid). Runs were more important to the buyer and financing more important to the seller. Negotiators could also choose whether to include a compatible issue, a second cartoon, which would provide gains to both parties if included. Finally, they could reach a contingent contract, based on the buyer and seller‘s differing expectations of the primary show‘s ratings. A contingent contract, for example, would require the seller to pay the buyer a rebate if the ratings fell below a certain, agreed-upon level. Procedures. All data collections followed the procedures outlined in Brett & Okumura (1998). Cartoon was participants‘ first negotiation exercise. Managers had no pre-course reading about deal-making negotiations. All had calculators. Managers received a standard introduction to their course and to the Cartoon exercise, which explained the roles of the two parties, and the three negotiable issues. Neither the second cartoon nor the contingent contract was mentioned. Managers prepared (60 minutes) with a same-role partner, but knew that they would negotiate as a solo, not a team. Buyers were assigned to sellers such that no two buyer preparation partners negotiated with two seller partners. Negotiating time (75 minutes) was strictly enforced. At the end, negotiators jointly completed a results sheet. They then individually

Culture, Trust, and Negotiation 16 completed a post-negotiation questionnaire, after which they received a standard debrief. The questionnaire response rate was high and comparable across cultures: overall, 90.15%. Data and Analysis. The independent variable, culture, was measured with the same questions as in Study 1. In this study, American culture was coded as 1 and Indian as 2. Appendix B contains the questions in the post-negotiation questionnaire. We measured trust that existed before (four questions; α = 0.86) and after (four similar questions; α = 0.92) the negotiation. When analyzing final trust, we controlled for initial trust. We also measured negotiation strategy. To develop these questions, we searched the literature on integrative and distributive negotiation strategies (e.g., Kimmel et al., 1980; Pruitt, 1981; Pruitt & Lewis, 1975; Weingart et al., 1990), ultimately constructing 12 self-report questions: six to measure Q&A and six to measure S&O. We factor analyzed the 12 questions, fitting two dimensions, which accounted for 44.42% of the common variance among the items. With Varimax rotation, they fit our a priori categories of Q&A and S&O; the reliabilities of the ensuing scales were α = 0.78 and α = 0.72, respectively. To measure insight, we followed Brett and Okumura (1998), asking how important price, runs, financing, and the second cartoon were to respondents and their counterparts. We constructed two measures of insight into tradeoffs. The first assessed whether negotiators correctly ascertained their counterparts‘ priorities (e.g., whether buyers indicated that financing was more important than runs for their seller counterparts). If negotiators assigned the counterpart‘s higher-priority issue a higher importance rating, we coded it 1. If they assigned it an equal or lower importance rating, we coded it 0, assuming that equal and incorrect ratings both indicated an absence of insight. The second measure of insight assessed whether negotiators correctly ascertained their own relative priorities (e.g., whether buyers indicated that runs were

Culture, Trust, and Negotiation 17 more important to themselves than to their counterparts). If negotiators gave a higher importance rating to the correct negotiator on both issues, we coded it 2; if ratings were correctly assigned for only one of the issues, we coded it 1; if ratings were incorrect for both issues, we coded it 0. We also calculated negotiators‘ joint gains (see Brett, 2007: 64-65), which in Cartoon, indicate Pareto optimality (Raiffa, 1982), such that any other agreement would generate a loss for one or both parties. Four impasses were included in the dataset; results were consistent when they were excluded, and the impasse rate did not differ by culture. Hypotheses 1, 2, 3, and 5 were proposed at the individual level of analysis. To control for the interdependence of dyad members and the risk of Type I error associated with biased standard errors, we tested these hypotheses with multilevel modeling (i.e., MLM; Raudenbush & Bryk, 2002). The first step in MLM is to determine whether the data has a group structure. If not, it is appropriate to analyze the data using standard OLS regression. Since our groups were negotiating dyads, a group structure reflecting dyad-level differences on the dependent variables indicates the need for MLM. To test for group structure, we first ran ICC1‘s on each dependent variable and then compared (using a -2 log likelihood test) a series of random intercept models that allowed dyad intercepts to vary against standard regression models that fixed the intercepts. Dyad membership explained a substantial portion of the variance in individual responses. The ICC1‘s for trust before (24.38%), Q&A (51.81%), S&O (53.27%), insight measure 1 (24.71%), insight measure 2 (42.26%) and trust after (50.81%) all differed significantly from zero, indicating a dyadic structure to our data. Comparison of the random intercept models and standard regression models indicated that the former better explained the interdependent nature of the data: trust before (p = .08), Q&A (p < 0.001), S&O (p < 0.001), insight measure 1 (p = 0.06), insight measure 2 (p = 0.001), and trust after (p < 0.001). The p-values were marginal for

Culture, Trust, and Negotiation 18 trust before and insight measure 1; however, this is not unusual for small groups like dyads (Bliese, 2000). To be as conservative as possible in controlling for interdependence, we used MLM to test all hypotheses. We ran a series of multilevel models predicting each of the dependent variables with culture, role, and the interaction between culture and role. Predictors were entered in raw-metric form. Role and role by culture were included as controls, to ensure none of our effects were rolespecific. Neither role as a main effect (average p = 0.52) nor role by culture (average p = 0.46) was significant in any MLM analysis. For ease of reference, the (n.s.) role main effects are listed in Table 1, and correlations are presented at the individual level. Hypothesis 4 was tested at the dyadic level, since both dyad members were from the same culture, and joint gains are only defined at the dyadic level. Results Overall the results of the MLM analyses (see Table 1) indicated good support for the theoretical model in Figure 1. Hypothesis 1 (Study 1), predicting that Indian negotiators would bring a low trust propensity into negotiation, was supported by the significant, negative coefficient on culture (β = -0.55, p = 0.03), indicating that Indian negotiators tended to trust less than American negotiators at the negotiation‘s outset. Hypothesis 2a, that American negotiators would report using Q&A more than Indian negotiators, was supported: Culture predicted Q&A (β = -1.22, p < 0.001), such that Indian negotiators reported using Q&A less than American negotiators. Likewise, Hypothesis 2b, that Indian negotiators would report using S&O more than American negotiators, was supported by the coefficient on culture (β = 0.83, p < 0.001). Hypothesis 3, that Indian negotiators would identify the relevant tradeoffs less accurately than American negotiators, was supported by both measures of insight. Culture significantly predicted

Culture, Trust, and Negotiation 19 the first measure of insight (β = -0.44, p < 0.001), indicating that Indian negotiators appreciated their counterparts‘ priorities less often than Americans. Indeed, only 31.48% of Indian negotiators correctly reported these priorities, while 75.38% of American negotiators did. Likewise, culture significantly predicted the second insight measure (β = -0.77, p < 0.001), indicating that Indian negotiators understood their relative priorities less often (correct 2/3 of the time). Hypothesis 4, predicting that Indians would negotiate lower joint gains than Americans, was supported. Indian managers (M = $3.43 million; SD = $1.26 million) achieved lower joint gains than American managers (M = $4.02 million; SD = $1.18 million), t(63) = 1.98, p = 0.05. Finally, Hypothesis 5, predicting that Indians would report less trust than Americans after the negotiation, was supported. Controlling for initial trust (β = 0.28, p = 0.01), culture still predicted post-negotiation trust (β = -0.55, p = 0.03), with Indian negotiators tending to report less trust after the negotiation than American negotiators. We did not run mediation analysis in Study 2. Study 2‘s questionnaire data were collected post-negotiation, making it difficult to justify a test of the causal order proposed in Figure 1 (James, Mulaik, & Brett, 1982). However, the correlations in Table 2 show that the variables were inter-correlated, in the directions suggested by our model in Figure 1. Culture was correlated significantly with all of the other variables. In addition, Q&A, and to a lesser degree S&O, was correlated with insight, and insight with joint gains. Trust after the negotiation linked to most other variables, especially culture, trust and strategy. Discussion Study 2 supported our model and hypotheses linking culture, trust, negotiation strategy, insight, and joint gains. As in Study 1, Indian negotiators were less apt to trust than American

Culture, Trust, and Negotiation 20 negotiators. Consistent with their self-reported trust, Indians reported engaging in less Q&A and more S&O than Americans. As predicted, these differences in strategy were associated with Indian negotiators‘ realization of fewer insights and lower joint gains than their American peers. Finally, Indian, as compared to American negotiators, appeared to leave the bargaining table with even lower trust than they brought to it. The results of Study 2 were consistent with our theorizing. They also showed that the cultural differences in trust propensity of MBA students (Study 1) were consistent with those of experienced executives. Finally, Study 2 showed a relationship between cultural propensity to trust and negotiation strategy. However, Study 2 could not fully establish strategy as the causal mechanism linking culture and joint gains. Study 3, which measures strategy-in-use, allows us to generalize from self-reported strategy to actual strategy. Study 3: Culture, Strategy-in-Use, and Outcomes Study 3 provides a stronger test of the causal implications depicted in Figure 1. It uses coded data reflecting negotiators‘ strategy-in-use to test the behavioral analogue of Hypothesis 2 (linking culture to strategy). It also provides further evidence for Hypothesis 3 (linking culture to insight) and 4 (linking culture to joint gains). Finally, it tests a new hypothesis implied by our model, indicating that strategy-in-use mediates the relationship between culture and joint gains. Two elements of Study 3‘s design contribute to the strength of its causal conclusions. First, coding negotiators‘ behaviors circumvents the biases inherent in self-report data (Weingart, Olekalns, & Smith, 2004). Second, the causal order in Study 3 (culture to strategy-in-use to joint gains) is clear. Negotiators‘ cultural background necessarily precedes their strategy-in-use. Furthermore, since strategies precede and even cause negotiation outcomes (Olekalns & Smith,

Culture, Trust, and Negotiation 21 2003; Weingart, Hyder, & Prietula, 1996; Weingart et al., 1990), and were measured from negotiation transcripts, strategies preceded joint gains. The behavioral analogue of Hypothesis 2, linking culture to strategy, is: Hypothesis 6a: American negotiators will engage in Q&A more often than Indian negotiators. Hypothesis 6b: Indian negotiators will engage in S&O more often than American negotiators. Our model implies that both Q&A and S&O strategies-in-use should mediate the relationship between culture and joint gains – albeit in opposite directions. (We chose this mediation because it encompassed the entire, causal chain in Figure 1; note that intermediate elements in a chain may be dropped and more distal links tested, so long as causal order is preserved (James, Mulaik, & Brett, 1982)). We expected American negotiators to generate higher joint gains via more Q&A and S&O. In contrast, we expected Indian negotiators to generate lower joint gains through more S&O and less Q&A: Hypothesis 7: Negotiation strategy-in-use will mediate the relationship between culture and joint gains. Methods Participants. The Study 3 sample was drawn from the same population as in Study 2. However, no one participated in both studies. The Study 3 Indian sample consisted of 25 dyads selected at random from a pool of 102 managers participating in one of several executive programs at a major Indian business school. The American sample consisted of 25 dyads selected at random from a pool of 186 managers participating in one of several executive MBA programs at a U.S. business school.

Culture, Trust, and Negotiation 22 The average age of the Indian managers was 46.35 (SD = 6.43), and the sample was 92.16% male. The average age of the Americans was 37.66 (SD = 4.82), and the sample was 76.92% male. Thus, the American sample had more females, χ21 = 4.55, p = 0.03, and was younger, t(99) = 7.68, p < 0.001, than the Indian sample. As in Study 2, we controlled all analyses for gender and age. Because Study 3 used a dyadic level of analysis, we operationalized the gender composition of the dyad as all male, versus male-female. Neither gender composition nor age had any effects. Simulation and Procedures. Study 3 used the same Cartoon simulation and procedures described in Study 2, except that all dyads consented to audio-record their negotiation. Each participant received a copy of his/her audio recording and a listening guide at the end of the course. Recordings were professionally transcribed, the Indian ones by Indian transcribers. Coding. We coded each speaking turn (all of one party‘s speech until ended by the beginning of the next party‘s speech) (Kimmel et al., 1980) for whether the negotiator speaking asked a question, conveyed an answer, substantiated, or made an offer. The literature on negotiation coding (Weingart et al., 2004) highlighted two additional elements (process comments and junk) that are commonly coded but unrelated to our hypotheses. Our coding scheme thus included six categories (see Appendix C). Each speaking turn in each transcript was allowed up to three codes. Junk was only coded when no other code was appropriate, No code was assigned more than once per speaking turn, and all speaking turns received at least one code (e.g., Kimmel et al., 1980; Weingart, Brett, Olekalns, & Smith, 2007). We hired three undergraduate coders, blind to the hypotheses and cross-cultural nature of our data. The coders were American. Because we randomly assigned transcripts to coders, however, any implicit cultural biases they might have had were randomly distributed across the

Culture, Trust, and Negotiation 23 American and Indian transcripts. During an intensive, two-month training, they independently coded over 10% of the transcripts and met five times to resolve disagreements through discussion. Throughout this period, we set aside random blocks of 451 speaking turns to assess coder reliability, reasoning that random blocks would best indicate reliability. By the end of the two-month period, at least two of the three coders agreed on over 70% of the codes assigned to these 451 turns. Cohen‘s Kappa for each pair of coders, not including the resolved turns, ranged from 0.74 to 0.77 and averaged 0.75 overall, indicating ―substantial‖ reliability (Landis & Koch, 1977). At this point, coders analyzed the remaining transcripts individually, although we intermingled shared transcripts periodically and checked for ongoing reliability. Data and Analysis. The level of analysis for Study 3 was the dyad. Culture and joint gains were operationalized as in Study 2. We also operationalized two different behavioral measures of insight – both at the dyadic level of analysis – reasoning that two, separate measures would enhance validity. The first measure, tradeoffs, indicated how much value the negotiators created through tradeoffs. This was operationalized by subtracting the price of the two cartoons from the agreement value (if negotiators did not include the second cartoon, we only subtracted the main cartoon‘s value). The second measure, special terms, indicated whether negotiators included the compatible issue and/or a contingent contract in their agreement. These outcomes – though not formal tradeoffs – require the same type of information exchange required for making tradeoffs (Brett, 2007). To operationalize Q& A and S&O, we used the percentage of codes (altogether 11,024 codes) in a given transcript that came from each category. To check the reliability of our measure, we also computed the percentage of all speaking turns (altogether 10,116 turns) in a transcript that included each category. These two measures were correlated at r > 0.9 and yielded

Culture, Trust, and Negotiation 24 similar results. Because percentage of codes appeared in previous research, captured the complexity of negotiators‘ statements, and accounted for potential cultural differences in wordiness, we report that measure below. Our final measure summed the relevant categories (e.g., Q&A=Q+A); results were identical for individual codes (e.g., Q, A). Results Study 3‘s results supported Study 2‘s findings and our model in Figure 1. Study 2‘s Hypothesis 4, predicting that Indian negotiators would achieve lower joint gains (M=$3.29 million, SD = $0.76 million) than American negotiators (M=$4.22 million, SD = $0.85 million), t(48) = 4.09, p < 0.001 was supported in Study 3. Our behavioral indices of insight help to explain this finding. Indian negotiators identified the tradeoff opportunities marginally less accurately than American negotiators, creating $2.89 million via tradeoffs (SD = $0.61 million), compared to the American negotiators‘ $3.26 million (SD = $0.74 million) in trade-offs, t(48) = 1.94, p = 0.06. Similarly, consistent with Study 2‘s Hypothesis 3, only 40% of Indian dyads included the compatible issue (the second cartoon) compared to 96% of American dyads, χ21 = 18.02, p < 0.001. Finally, none of the Indian dyads, compared to 16% of the American dyads, capitalized on buyers‘ and sellers‘ differing expectations by creating a contingent contract, χ21 = 4.35, p = 0.04. Thus, whenever negotiators could create mutually-beneficial tradeoffs, Americans not only identified the opportunities but acted upon them more often than Indians. Culture was related to strategies-in-use, and strategies-in-use to joint gains. Hypothesis 6a, that Americans would use Q&A more than Indians, was supported: 54.57% (SD = 12.77) of American codes and 33.91% (SD = 11.27) of Indian codes were Q&A, t(48) = 6.07, p < 0.001. Hypothesis 6b, that Indians would use S&O more than Americans, was also supported: 59.39%

Culture, Trust, and Negotiation 25 (SD = 11.72) of Indian codes and 34.70% (10.94) of American codes were S&O, t(48) = 7.70, p < 0.001. Hypothesis 7 predicted that strategy-in-use would mediate the relationship between culture and joint gains. The correlations (see Table 3) and regressions provided initial support: both of the strategies, as well as their underlying behaviors, were related to culture and joint gains. Examining Q&A first, both culture and Q&A were significant predictors of joint gains; when both were included as predictors, culture became non-significant, suggesting full mediation. A bootstrap analysis (Shrout & Bolger, 2002) supported mediation of the culture-joint gains relationship by Q&A: the 95% CI (-892,396 to -61,678) did not include zero, demonstrating significant mediation. Examining S&O, both culture and S&O were significant predictors of joint gains; when both were included in the regression, culture again became nonsignificant, suggesting full mediation. A second bootstrap analysis produced a 95% CI of (1,167,715 to -162,929), demonstrating significant mediation (see Figure 2). Overall, both Q&A and S&O independently mediated the relationship between culture and joint gains. Q&A mediated by facilitating joint gains (especially for American negotiators), and S&O mediated by undermining joint gains (especially for Indian negotiators). Discussion Study 3 illustrated how a lack of insight into the other party‘s priorities (as identified in Study 2) translates into low joint gains. Compared to American negotiators, Indian negotiators in Study 3 made fewer tradeoffs, acted upon compatible issues less often, and reached fewer contingent contracts. Taken together, these features of Indians‘ agreements yielded low joint gains. Analysis of negotiators‘ strategy-in-use further demonstrated that Indians‘ dominant use of S&O, and less frequent use of Q&A, undermined their joint gains – consistent with Study 2. By

Culture, Trust, and Negotiation 26 comparison to Indian negotiators, American negotiators used Q&A frequently and S&O infrequently, which generated enhanced insight into tradeoff opportunities and joint gains. General Discussion Three studies documented that Indian and American negotiators bring different trust propensities to the negotiating table, with far-reaching consequences for their negotiation strategy, insight into tradeoff opportunities, and joint gains. Relative to the American negotiators in our studies, Indian negotiators assumed little trust, used S&O strategy more, used Q&A strategy less, achieved fewer insights, and walked away with lower gains and reduced trust. Our results contribute to theory concerning the role of culture, trust, and strategy in negotiation. They also offer an immediate explanation for the divergent performance of Indian and American negotiators. We first address contributions to theory. We then consider how to help negotiators who bring low trust into negotiation (for cultural or other reasons) to achieve joint gains. Finally, we discuss the studies‘ strengths and limitations, along with opportunities for further research. Implications for Theory An important theoretical contribution of these studies is evidence for the series of relationships depicted in Figure 1. We found that when culture endows people with a low versus high propensity to trust in negotiations, they deploy different negotiation strategies, which account for cultural differences in their joint gains. These findings document a new theoretical chain that can inform research on culture, trust, and negotiation and generate predictions about intercultural negotiations. Extant negotiation research generally treats trust as an individual difference or even a contextual factor, but not a cultural propensity (e.g., Butler, 1999; Olekalns et al., 2007).

Culture, Trust, and Negotiation 27 Furthermore, with the exception of behavioral game theory studies (e.g., Bottom, Gibson, Daniels, & Murnighan, 2002), there is little research documenting an empirical relationship between trust and negotiation strategy (Thompson, Wang, & Gunia, 2010). The research that does document such a link (Butler, 1995, 1999) generally calls for examination of the antecedents and mechanisms of trust in negotiations. Our research identifies culture as an antecedent of trust in negotiation, and shows that differences in negotiation strategy are associated with differences in trust. We also provide a theoretical explanation for the links between high trust and Q&A strategy on one hand, and low trust and S&O strategy on the other. To reiterate, Q&A requires trust because it leaves the negotiator vulnerable to exploitation. S&O does not require trust because it reveals little about the negotiator and resonates with fixed-pie expectations. Furthermore, Brett (2007) points out that S&O minimizes the risk of falling prey to a lie, as counterparts usually do not make offers at-odds with their own self-interest. Thus, because they wish to avoid the vulnerability inherent in sharing or receiving information, negotiators who do not trust do not engage in Q&A. Instead, they turn to the lower-risk, but potentially lower-reward strategy of S&O. Another contribution of the current studies stems directly from the evidence that cultural differences in trust are associated with large differences in outcomes. The current studies suggest links between micro and macro-level phenomena: On a micro level, our findings parallel Yamagichi‘s research linking cultural differences in trust to decision making (Yamagishi & Yamagishi, 1994). On a macro level, they parallel socio-political research linking trust to country-level outcomes like prosperity (e.g., Fukuyama, 1995). Our findings justify the concerns we raised at the outset of the paper, that low trust propensity in negotiations may hamper Indian managers negotiating on the global stage.

Culture, Trust, and Negotiation 28 Culture by no means completely determines negotiation strategy. The correlations in Studies 2 and 3 show that negotiators from both Indian and American cultures negotiated higher joint gains to the extent that they used Q&A, and lower joint gains to the extent that they used S&O. The significant mediation analysis in Study 3 suggests that strategy, not culture, ultimately provides the proximal explanation for variable joint gains. Cultural differences in negotiation outcomes arise from strategies, born in part of cultural beliefs about trust in negotiations. Opportunities for Future Research Our studies confirm the prior finding that Americans use Q&A to negotiate joint gains. However, ours are the first to document that the relatively low joint gains realized by Indian negotiators (Brett, 2007) are due to an emphasis on S&O strategy, cued by low trust in negotiations. These findings are consistent with prior American research showing low joint gains from issue-by-issue haggling (Pruitt & Lewis, 1975; Weingart et al., 1990; Adair et al. 2007). However, other negotiators from relatively low-trust cultures (e.g., the Japanese; Yamagishi &Yamagishi, 1994) seem to use S&O strategy to better effect (Adair et al, 2001; 2007). For example, Japanese negotiators generate insight and joint gains comparable to American negotiators (Brett & Okumura, 1998). Wherein lies the difference between Indian and Japanese negotiators use of S&O? We propose that future research should investigate the two cultures‘ use of linear versus holistic modes of thinking (Nisbett et al., 2001), with the answer to this enigma lying at the nexus of trust and thought. Implications for Practice The practical question is how Indian and other negotiators tending toward low trust can nevertheless negotiate high joint gains. We suggest several interventions that may help negotiators who bring a cultural or personal propensity for low trust into negotiations to achieve

Culture, Trust, and Negotiation 29 joint gains. Even negotiators with a propensity to trust may find these prescriptions useful when negotiating with low-trust counterparts. It is important to acknowledge that interventions to build trust in negotiations seem unrealistic: Low-trust assumptions appear deep-seated, and are likely to be functional, in cultures like India (The Economist, 12/11/08). Training low-trust negotiators to change their propensity, if possible and ethical, would only work so long as each of their counterparts was also trained and did not defect from the training. One serious trust violation – a near certainty in countries of millions or billions of negotiators – could outweigh any exhortations to trust (Bottom et al., 2002; Kim, Ferrin, Cooper, & Dirks, 2004; Olekalns, Brett, & Weingart, 2003). Interventions that acknowledge low trust and the associated preference for offer-making may provide a better solution. For example, low-trust negotiators may learn to reach insights from offers – especially multiple-issue offers (Brett, 2007; Medvec & Galinsky, 2005). There are several reasons for our optimism. First, research suggests that multiple-issue offers lead to higher joint gains than a sequence of single-issue offers (e.g., Hyder, Prietula, & Weingart, 2000; Kelley, 1966; Mannix, Thompson, & Bazerman, 1989; Olekalns & Smith, 2003; Weingart, Bennett, & Brett, 1993). Negotiators can ―stumble‖ into mutually-beneficial tradeoffs by cycling through multiple-issue offers, whereas stumbling into tradeoffs through a litany of single-issue offers can be difficult (Kimmel et al., 1980; Pruitt, 1981). Second, low-trust negotiators should be comfortable with offers because they reveal only a modicum of truth about priorities: rarely do offers shed full light on negotiators‘ interests, but rarely are such offers completely at-odds with negotiators‘ interests either (Brett, 2007). Thus, although offers reveal more about positions than interests, they at least provide directional information on interests. Whereas low-trust negotiators might not believe the information

Culture, Trust, and Negotiation 30 revealed through Q&A, nor wish to reciprocate it by answering the counterpart‘s questions, they may believe offers and reciprocate with counteroffers, and perhaps substantiation. By examining the pattern of offers, counter offers, and substantiation, negotiators may be able to infer something about their counterpart‘s priorities (Brett, 2007). Third, studies show that negotiators can learn to use particular negotiation strategies (e.g., contingent contracts) via analogical learning, a process of accumulating and comparing examples to appreciate abstract learning principles (Thompson, Gentner, & Loewenstein, 2000). Although we know of no studies using analogical learning to teach inferences from offers, it may be a promising approach. Another approach lies in the assignment of negotiation goals: Giving negotiators mastery learning goals has helped teams to improve their joint gains via logrolling (Bereby-Meyer, Moran, & Unger-Aviram, 2004). Whether with analogical learning or goal assignment, it seems feasible to help Indian and other low-trust negotiators turn the favored S&O strategy to their advantage, by helping them learn to draw inferences from multiple-issue offers. Without such an intervention, Indian and low-trust managers, negotiating globally, may face formidable obstacles in realizing joint gains. Furthermore, these obstacles are likely to surface with both Western and Eastern counterparts. When faced with the S&O strategy, for example, Westerners may quickly abandon their favored Q&A strategy and reciprocate S&O (Johnson & Cullen, 2002), spiraling away from agreements and joint gains (Brett, 2007; Brett, Shapiro, & Lytle, 1998; Weingart et al., 2007). Likewise, if East Asians draw upon their high context communication skills and a penchant for holistic thinking (Nisbett, et al., 2001), they may reach inferences about the Indian negotiator‘s preferences and priorities from S&O (Brett, 2007; Adair et al, 2007) and take advantage. Thus, Indian negotiators with an East Asian counterpart may realize low individual, as well as joint gains.

Culture, Trust, and Negotiation 31 Study Strengths and Limitations Our three studies have several methodological and analytical strengths. These strengths support generalizability and strong inference. The studies featured three datasets – all of which drew independent samples from similar populations. Unsurprisingly, demographics (e.g., gender, age) bore little relationship to negotiators‘ trust, strategy, and outcomes, but culture influenced all three. The consistency of our findings across three, relatively diverse samples of Indian and American negotiators increases our confidence in the generalizability of our findings to the managerial populations negotiating business agreements in these two cultures. At the same time, we recognize that our findings highlight central tendencies in a subset of the populations in the cultures under investigation. Certainly, our samples from pools of welleducated and experienced managers do not reflect the full populations of either country. Furthermore, even when central tendencies reflect large and significant mean differences, there are always outliers whose experience and worldview allow them to transcend cultural tendencies. Nevertheless, the existence of large and significant differences, especially in the Study 3 negotiation strategies-in-use, endows the findings with credibility. Our studies also used a variety of methods and measures – survey and simulation, selfreport and coded strategy – to measure key variables: beliefs, behaviors, and outcomes. Regardless of method or measure, the data supported the same causal chain from culture, to trust propensity, to negotiation strategy, to insight, to joint gains. Although no single study addressed all links in the causal chain simultaneously, each addressed an overlapping portion of the model. In particular, the similarity of the Study 2 and 3 methods allowed us to triangulate upon the relationship between culture, trust, strategy, insight, and outcomes. These studies‘ outcome data, negotiated joint gains, also revealed the real and consequential implications of trust propensity.

Culture, Trust, and Negotiation 32 Our studies allowed us to draw some reasonably strong inferences about the causality of culture in dictating trust propensity and negotiation strategy – and the causality of strategy in dictating insights and joint gains. Because culture was antecedent to all of our measures, we were able to identify trust as a culture-relevant predictor of strategy, and strategy as a trust-relevant predictor of insight and joint gains. Bootstrapped mediation analyses supported the causal inferences proposed by our model. Finally, the consistency of the strategy-outcome results across Study 2‘s self-report and Study 3‘s behavioral data demonstrated substantial validity. The intra-cultural, comparative nature of our studies limited us from making empirical, as opposed to theoretical, generalizations about the implications of our findings for intercultural negotiations. Since prior research has documented the poor fit between negotiation strategies born of high and low context communication cultures (Adair & Brett, 2005; Brett et al., 1998), we suggest that intercultural negotiations may prove challenging for Indian negotiators and their counterparts – especially in light of trust propensity. However, this proposition is clearly open to empirical test in future research. There is a dearth of inter-cultural negotiation research in general. Our theoretical model of the relationships between culture, trust, strategy, insight, and joint gains provides a solid platform for future research investigating the fit between different cultural beliefs and negotiation strategies. Of high priority is theorizing about factors affecting negotiations between Indians and East Asians. In addition, with so much negotiation training available around the world, it is imperative to determine whether negotiators with a low propensity to trust can learn to make integrative tradeoffs if trained with analogical learning or mastery goals. Conclusion

Culture, Trust, and Negotiation 33 The three studies reported in this paper deepen our understanding of culture‘s impact on negotiation, via trust and strategy. With these studies, we have causal evidence that culture promotes higher or lower trust, with material and substantial consequences for the negotiation that follows.

Culture, Trust, and Negotiation 34 References Adair, W. L., & Brett, J. M. (2005). The negotiation dance: Time, culture, and behavioral sequences in negotiation. Organization Science, 16(1), 33-51. Adair, W. L., Weingart, L. R., & Brett, J. M. (2007). The timing and function of offers in U.S. and Japanese negotiations. Journal of Applied Psychology, 92(4), 1056-1068. Allinson, C. W., & Hayes, J. (1988). Cultural differences in the learning styles of managers. Management International Review, 28(3), 75-78. Bazerman, M. H., & Neale, M. A. (1992). Negotiating rationally. New York: Free Press. Bereby-Meyer, Y., Moran, S., Unger-Aviram, E. (2004). When performance goals deter performance: Transfer of skills in integrative negotiations. Organizational Behavior and Human Decision Processes, 93, 142-154. Bliese, P. D. (2000) Within-group agreement, non-independence, and reliability: Implications for data aggregation and analysis. In K. J. Klein & S. W. Kozlowski (Eds.), Multilevel Theory, Research, and Methods in Organizations (pp. 349-381). San Francisco: JosseyBass, Inc. Bottom, W. P., Gibson, K., Daniels, S. E., & Murnighan, J. K. (2002). When talk is not cheap: Substantive penance and expressions of intent in rebuilding cooperation. Organization Science, 13(5), 497-513. Branzei, O., Vertinsky, I., & Camp, R. D. (2007). Culture-contingent signs of trust in emergent relationships. Organizational Behavior and Human Decision Processes, 104(1), 61-82. Brett, J. M. (2007). Negotiating globally: How to negotiate deals, resolve disputes, and make decisions across cultural boundaries (2nd ed.). San Francisco: Jossey-Bass.

Culture, Trust, and Negotiation 35 Brett, J. M., & Okumura, T. (1998). Inter- and intracultural negotiation: U.S. and Japanese negotiators. Academy of Management Journal, 41(5), 495-510. Brett, J. M., Shapiro, D. L., & Lytle, A. L. (1998). Breaking the bonds of reciprocity in negotiations. Academy of Management Journal, 41(4), 410-424. Buchan, N. R., Croson, R. T. A., & Dawes, R. M. (2002). Swift neighbors and persistent strangers: A cross-cultural investigation of trust and reciprocity in social exchange. American Journal of Sociology, 108, 168-206. Butler, J. K. (1995). Behaviors, trust, and goal achievement in a win-win negotiating role play. Group & Organization Management, 20(4), 486-501. Butler, J. K. (1999). Trust expectations, information sharing, climate of trust, and negotiation effectiveness and efficiency. Group & Organization Management, 24(2), 217-238. Colquitt, J. A., Scott, B. A., & LePine, J. A. (2007). Trust, trustworthiness, and trust propensity: A meta-analytic test of their unique relationships with risk taking and job performance. Journal of Applied Psychology, 92(4), 909-927. Delhey, J., & Newton, K. (2005). Predicting cross-national levels of social trust: Global pattern or Nordic exceptionalism? European Sociological Review, 21(4), 311-327. Deutsch, M. (1973). The resolution of conflict: Constructive and destructive processes. New Haven, CT: Yale University Press. Dirks, K. T., Lewicki, R. J., & Zaheer, A. (2009). Repairing relationships within and between organizations: Building a conceptual foundation. Academy of Management Review, 34(1), 68-84. Dispute Resolution Research Center (2008). Evanston, IL: Northwestern University.

Culture, Trust, and Negotiation 36 Druckman, D., Benton, A. A., Ali, F., & Bagur, J. S. (1976). Cultural differences in bargaining behavior: India, Argentina, and United States. Journal of Conflict Resolution, 20(3), 413452. The Economist. (1/11/08). An elephant, not a tiger: A special report on India. Retrieved 4/1/09, from http://www.economist.com/specialreports/ displaystory.cfm?story_id=12749735. Fang, T. (1999). Chinese business negotiating style. Thousand Oaks, CA: Sage Publications. Ferrin, D. L., & Gillespie, N. (2010). Trust differences across national-societal cultures: Much to do, or much ado about nothing? In M. Saunders, D. Skinner, G. Dietz, N. Gillespie & R. J. Lewicki (Eds.), Trust Across Cultures: Theory and Practice. Cambridge, UK: Cambridge University Press. Fisher, R., & Ury, W. (1981). Getting to yes: Negotiating agreement without giving in. Boston: Houghton Mifflin. Fukuyama, F. (1995). Trust: The social virtues and the creation of prosperity. New York: Free Press. Glenn, E. S., & Glenn, C. G. (1981). Man and mankind: Conflict and communication between cultures. Norwood, NJ: Ablex Publishing Corporation. Gouldner, A. W. (1960). The norm of reciprocity: A preliminary statement. American Sociological Review, 25(2), 161-178. Hall, E. T. (1976). Beyond culture (1st ed.). Garden City, NJ: Anchor Press. Hayes, J., & Allinson, C. W. (2000). Cross-national differences in cognitive style: Implications for management. International Journal of Human Resource Management, 11(1), 161170.

Culture, Trust, and Negotiation 37 Huff, L. & Kelley, L. (2003). Levels of organizational trust in individualist versus collectivist societies: A seven-nation study. Organization Science, 14, 81-90. Hyder, E. B., Prietula, M. J., & Weingart, L. R. (2000). Getting to best: Efficiency versus optimality in negotiation. Cognitive Science, 24(2), 169-204. Inglehart, R., Basáñez, M., & Menéndez Moreno, A. (1998). Human Values and Beliefs: A Cross-cultural Sourcebook: Political, Religious, Sexual, and Economic Norms in 43 Societies; Findings from the 1990-1993 World Values Survey. Ann Arbor, MI: University of Michigan Press. James, L. R., Mulaik, S. A., & Brett, J. M. (1982). Causal analysis: Assumptions, models, and data. Beverly Hills, CA: Sage. Johnson, J. L., & Cullen, J. B. (2002). Trust in cross-cultural relationships. In M. J. Gannon & K. Newman (Eds.), The Blackwell Handbook of Cross-Cultural Management. Oxford, UK: Blackwell Publishing. Kapoor, S., Hughes, P. C., Baldwin, J. R., & Blue, J. (2003). The relationship of individualismcollectivism and self-construals to communication styles in India and the United States. International Journal of Intercultural Relations, 27, 683-700. Kelley, H. H. (1966). A Classroom Study of Dilemmas in Interpersonal Negotiation. In K. Archibald (Ed.), Strategic Interaction and Conflict: Institute of International Studies, University of California. Kim, P. H., Ferrin, D. L., Cooper, C. D., & Dirks, K. T. (2004). Removing the shadow of suspicion: The effects of apology versus denial for repairing competence- versus integrity-based trust violations. Journal of Applied Psychology, 89(1), 104-118.

Culture, Trust, and Negotiation 38 Kimmel, M. J., Pruitt, D. G., Magenau, J. M., Konargoldband, E., & Carnevale, P. J. D. (1980). Effects of trust, aspiration, and gender on negotiation tactics. Journal of Personality and Social Psychology, 38(1), 9-22. Kumar, R., & Worm, V. (2004). Institutional dynamics and the negotiation process: Comparing India and China. International Journal of Conflict Management, 15(3), 304-334. Landis, J. R., & Koch, G. G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33(1), 159-174. Lewicki, R. J., Stevenson, M. A., & Bunker, B. B. (1997). The three components of interpersonal trust: Instrument development and differences across relationships. Paper presented at the Academy of Management meeting. Mannix, E. A., Thompson, L. L., & Bazerman, M. H. (1989). Negotiation in small groups. Journal of Applied Psychology, 74(3), 508-517. March, R. M. (1988). The Japanese negotiator: Subtlety and strategy beyond western logic (1st ed.). Tokyo: Kodansha International. Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. Academy of Management Review, 20(3), 709-734. Medvec, V., & Galinsky, A. D. (2005). Putting more on the table: How making multiple offers can increase the final value of the deal. HBS Negotiation Newsletter, 4. Meyerson, D., Weick, K. E., & Kramer, R. M. (1996). Swift trust and temporary groups. In R. M. Kramer & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research (pp. 166-195). Thousand Oaks, CA: Sage. Nisbett, R. E., Peng, K. P., Choi, I., & Norenzayan, A. (2001). Culture and systems of thought: Holistic versus analytic cognition. Psychological Review, 108(2), 291-310.

Culture, Trust, and Negotiation 39 OECD Policy Brief. (2007). Economic Survey of India, 2007. 10, 1-12. Olekalns, M., Brett, J. M., & Weingart, L. R. (2003). Phases, transitions and interruptions: Modeling processes in multi-party negotiations. International Journal of Conflict Management, 14(3), 191-211. Olekalns, M., Lau, F., & Smith, P. L. (2007). Resolving the empty core: Trust as a determinant of outcomes in three-party negotiations. Group Decision and Negotiation, 16(6), 527538. Olekalns, M., & Smith, P. L. (2003). Testing the relationships among negotiators' motivational orientations, strategy choices, and outcomes. Journal of Experimental Social Psychology, 39(2), 101-117. Olekalns, M., & Smith, P. L. (2005). Moments in time: Metacognition, trust, and outcomes in dyadic negotiations. Personality and Social Psychology Bulletin, 31(12), 1696-1707. Parikh, J., Neubauer, F., & Lank, A. G. (1994). Intuition: The new frontier of management. Oxford: Blackwell. Pruitt, D. G. (1981). Negotiation behavior. New York: Academic Press. Pruitt, D. G., & Lewis, S. A. (1975). Development of integrative solutions in bilateral negotiation. Journal of Personality and Social Psychology, 31(4), 621-633. Raiffa, H. (1982). The art and science of negotiation. Cambridge, MA: Belknap Press of Harvard University Press. Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (Second ed.). Thousand Oaks, CA: Sage Publications. Requejo, W. H., & Graham, J. H. (2008). Global negotiation: The new rules. New York: Palgrave MacMillan.

Culture, Trust, and Negotiation 40 Reuters. (1/29/08). India's outsourcing revenue at $50 bln by 2012-study. Retrieved 4/1/09. from http://www.reuters.com/article/rbssTechMediaTelecomNews/ idUSBOM14477620080129. Rousseau, D. M., Sitkin, S. B., Burt, R. S., & Camerer, C. (1998). Not so different after all: A cross-discipline view of trust. Academy of Management Review, 23(3), 393-404. Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: New procedures and recommendations. Psychological Methods, 7(4), 422-445. Thompson, L. L., Gentner, D., & Loewenstein, J. (2000). Avoiding missed opportunities in managerial life: Analogical training more powerful than individual case training. Organizational Behavior and Human Decision Processes, 82(1), 60-75. Thompson, L. L., Wang, J., & Gunia, B. C. (2010). Negotiation. Annual Review of Psychology, 61, 491-515. Transparency International. (2009). Corruption Perceptions Index 2009. Retrieved 12/15/09, from http://www.transparency.org/policy_research/surveys_indices/cpi. United Nations Development Programme (UNDP). (2008). 2007/2008 Human Development Report. Retrieved 4/1/09 from http://hdrstats.undp.org/indicators/24.html. U.S. Census Bureau. (2009). Foreign Trade Statistics. Retrieved 4/1/09, from http://www.census.gov/foreign-trade/balance/c5330.html. Walton, R. E., & McKersie, R. B. (1965). A behavioral theory of labor negotiations. New York: McGraw-Hill. Weber, J. M., Malhotra, D., & Murnighan, J. K. (2005). Normal acts of irrational trust: Motivated attributions and the trust development process. Research in Organizational Behavior: An Annual Series of Analytical Essays and Critical Reviews, Vol 26, 75-101.

Culture, Trust, and Negotiation 41 Weingart, L. R., Bennett, R. J., & Brett, J. M. (1993). The impact of consideration of issues and motivational orientation on group negotiation process and outcome. Journal of Applied Psychology, 78(3), 504-517. Weingart, L. R., Brett, J. M., Olekalns, M., & Smith, P. L. (2007). Conflicting social motives in negotiating groups. Journal of Personality and Social Psychology, 93(6), 994-1010. Weingart, L. R., Hyder, E. B., & Prietula, M. J. (1996). Knowledge matters: The effect of tactical descriptions on negotiation behavior and outcome. Journal of Personality and Social Psychology, 70, 1205-1217. Weingart, L. R., Olekalns, M., & Smith, P. L. (2004). Quantitative coding of negotiation strategies. International Negotiation, 9, 441-446. Weingart, L. R., Thompson, L. L., Bazerman, M. H., & Carroll, J. S. (1990). Tactical behavior and negotiation outcomes. International Journal of Conflict Management, 1(1), 7-31. Yamagishi, T., & Yamagishi, M. (1994). Trust and commitment in the United States and Japan. Motivation and Emotion, 18(2), 129-166. Zand, D. E. (1972). Trust and managerial problem solving. Administrative Science Quarterly, 17(2), 229-239.

Culture, Trust, and Negotiation 42 Figure 1: Model

Culture

Trust Propensity

Negotiation Strategies

Insights

Joint Gains

Culture, Trust, and Negotiation 43 Figure 2: Study 3 Mediation (coefficients standardized) Q&A 0.53*** (0.34*)

-0.66***

Nationality 1=American, 2=Indian

-0.51*** (-0.28ns)

Joint Gains

S&O -0.58*** (-0.44*)

0.74***

Nationality 1=American, 2=Indian

-0.51*** (-0.18ns)

Joint Gains

Culture, Trust, and Negotiation 44 Table 1: Study 2 MLM Results (multiple levels) Dependent Variable Trust Before

Predictors

β

S.E.

t

Intercept (fixed) 5.89 0.49 12.10 Culture -0.55 0.25 -2.19 Role 0.14 0.20 0.69 Q&A Intercept (fixed) 6.99 0.42 16.61 Culture -1.22 0.23 -5.36 Role -0.10 0.16 -0.66 S&O Intercept (fixed) 2.58 0.42 6.14 Culture 0.83 0.23 3.54 Role -0.20 0.14 -1.41 Insight Intercept (fixed) 1.18 0.18 6.62 Measure 1 Culture -0.44 0.09 -5.09 Role 0.01 0.08 0.14 Insight Intercept (fixed) 2.28 0.29 7.92 Measure 2 Culture -0.77 0.15 -5.09 Role -0.08 0.11 -0.71 Trust Intercept (fixed) 5.05 0.84 6.00 After Culture -0.95 0.33 -2.90 Trust Before 0.28 0.10 2.68 Role -0.10 0.20 -0.52 Culture (1=US, 2=India), Role (1=Buyer, 2=Seller)

p